[go: up one dir, main page]

EP3559841B1 - Normalisation de couverture de base et son utilisation pour détecter une variation du nombre de copies - Google Patents

Normalisation de couverture de base et son utilisation pour détecter une variation du nombre de copies

Info

Publication number
EP3559841B1
EP3559841B1 EP17838107.5A EP17838107A EP3559841B1 EP 3559841 B1 EP3559841 B1 EP 3559841B1 EP 17838107 A EP17838107 A EP 17838107A EP 3559841 B1 EP3559841 B1 EP 3559841B1
Authority
EP
European Patent Office
Prior art keywords
base
gene
coverage
level
region
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
EP17838107.5A
Other languages
German (de)
English (en)
Other versions
EP3559841A1 (fr
Inventor
Catalin Barbacioru
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Grail Inc
Original Assignee
Grail Inc
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Grail Inc filed Critical Grail Inc
Publication of EP3559841A1 publication Critical patent/EP3559841A1/fr
Application granted granted Critical
Publication of EP3559841B1 publication Critical patent/EP3559841B1/fr
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • C—CHEMISTRY; METALLURGY
    • C12—BIOCHEMISTRY; BEER; SPIRITS; WINE; VINEGAR; MICROBIOLOGY; ENZYMOLOGY; MUTATION OR GENETIC ENGINEERING
    • C12Q—MEASURING OR TESTING PROCESSES INVOLVING ENZYMES, NUCLEIC ACIDS OR MICROORGANISMS; COMPOSITIONS OR TEST PAPERS THEREFOR; PROCESSES OF PREPARING SUCH COMPOSITIONS; CONDITION-RESPONSIVE CONTROL IN MICROBIOLOGICAL OR ENZYMOLOGICAL PROCESSES
    • C12Q1/00—Measuring or testing processes involving enzymes, nucleic acids or microorganisms; Compositions therefor; Processes of preparing such compositions
    • C12Q1/68—Measuring or testing processes involving enzymes, nucleic acids or microorganisms; Compositions therefor; Processes of preparing such compositions involving nucleic acids
    • C12Q1/6869—Methods for sequencing
    • C—CHEMISTRY; METALLURGY
    • C12—BIOCHEMISTRY; BEER; SPIRITS; WINE; VINEGAR; MICROBIOLOGY; ENZYMOLOGY; MUTATION OR GENETIC ENGINEERING
    • C12Q—MEASURING OR TESTING PROCESSES INVOLVING ENZYMES, NUCLEIC ACIDS OR MICROORGANISMS; COMPOSITIONS OR TEST PAPERS THEREFOR; PROCESSES OF PREPARING SUCH COMPOSITIONS; CONDITION-RESPONSIVE CONTROL IN MICROBIOLOGICAL OR ENZYMOLOGICAL PROCESSES
    • C12Q1/00—Measuring or testing processes involving enzymes, nucleic acids or microorganisms; Compositions therefor; Processes of preparing such compositions
    • C12Q1/68—Measuring or testing processes involving enzymes, nucleic acids or microorganisms; Compositions therefor; Processes of preparing such compositions involving nucleic acids
    • C12Q1/6869—Methods for sequencing
    • C12Q1/6874—Methods for sequencing involving nucleic acid arrays, e.g. sequencing by hybridisation
    • C—CHEMISTRY; METALLURGY
    • C12—BIOCHEMISTRY; BEER; SPIRITS; WINE; VINEGAR; MICROBIOLOGY; ENZYMOLOGY; MUTATION OR GENETIC ENGINEERING
    • C12Q—MEASURING OR TESTING PROCESSES INVOLVING ENZYMES, NUCLEIC ACIDS OR MICROORGANISMS; COMPOSITIONS OR TEST PAPERS THEREFOR; PROCESSES OF PREPARING SUCH COMPOSITIONS; CONDITION-RESPONSIVE CONTROL IN MICROBIOLOGICAL OR ENZYMOLOGICAL PROCESSES
    • C12Q1/00—Measuring or testing processes involving enzymes, nucleic acids or microorganisms; Compositions therefor; Processes of preparing such compositions
    • C12Q1/68—Measuring or testing processes involving enzymes, nucleic acids or microorganisms; Compositions therefor; Processes of preparing such compositions involving nucleic acids
    • C12Q1/6876—Nucleic acid products used in the analysis of nucleic acids, e.g. primers or probes
    • C12Q1/6883—Nucleic acid products used in the analysis of nucleic acids, e.g. primers or probes for diseases caused by alterations of genetic material
    • C12Q1/6886—Nucleic acid products used in the analysis of nucleic acids, e.g. primers or probes for diseases caused by alterations of genetic material for cancer
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16B—BIOINFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR GENETIC OR PROTEIN-RELATED DATA PROCESSING IN COMPUTATIONAL MOLECULAR BIOLOGY
    • G16B20/00—ICT specially adapted for functional genomics or proteomics, e.g. genotype-phenotype associations
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16B—BIOINFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR GENETIC OR PROTEIN-RELATED DATA PROCESSING IN COMPUTATIONAL MOLECULAR BIOLOGY
    • G16B20/00—ICT specially adapted for functional genomics or proteomics, e.g. genotype-phenotype associations
    • G16B20/20—Allele or variant detection, e.g. single nucleotide polymorphism [SNP] detection
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16B—BIOINFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR GENETIC OR PROTEIN-RELATED DATA PROCESSING IN COMPUTATIONAL MOLECULAR BIOLOGY
    • G16B30/00—ICT specially adapted for sequence analysis involving nucleotides or amino acids
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16B—BIOINFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR GENETIC OR PROTEIN-RELATED DATA PROCESSING IN COMPUTATIONAL MOLECULAR BIOLOGY
    • G16B30/00—ICT specially adapted for sequence analysis involving nucleotides or amino acids
    • G16B30/10—Sequence alignment; Homology search
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16B—BIOINFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR GENETIC OR PROTEIN-RELATED DATA PROCESSING IN COMPUTATIONAL MOLECULAR BIOLOGY
    • G16B40/00—ICT specially adapted for biostatistics; ICT specially adapted for bioinformatics-related machine learning or data mining, e.g. knowledge discovery or pattern finding
    • C—CHEMISTRY; METALLURGY
    • C12—BIOCHEMISTRY; BEER; SPIRITS; WINE; VINEGAR; MICROBIOLOGY; ENZYMOLOGY; MUTATION OR GENETIC ENGINEERING
    • C12Q—MEASURING OR TESTING PROCESSES INVOLVING ENZYMES, NUCLEIC ACIDS OR MICROORGANISMS; COMPOSITIONS OR TEST PAPERS THEREFOR; PROCESSES OF PREPARING SUCH COMPOSITIONS; CONDITION-RESPONSIVE CONTROL IN MICROBIOLOGICAL OR ENZYMOLOGICAL PROCESSES
    • C12Q2600/00—Oligonucleotides characterized by their use
    • C12Q2600/112—Disease subtyping, staging or classification
    • C—CHEMISTRY; METALLURGY
    • C12—BIOCHEMISTRY; BEER; SPIRITS; WINE; VINEGAR; MICROBIOLOGY; ENZYMOLOGY; MUTATION OR GENETIC ENGINEERING
    • C12Q—MEASURING OR TESTING PROCESSES INVOLVING ENZYMES, NUCLEIC ACIDS OR MICROORGANISMS; COMPOSITIONS OR TEST PAPERS THEREFOR; PROCESSES OF PREPARING SUCH COMPOSITIONS; CONDITION-RESPONSIVE CONTROL IN MICROBIOLOGICAL OR ENZYMOLOGICAL PROCESSES
    • C12Q2600/00—Oligonucleotides characterized by their use
    • C12Q2600/118—Prognosis of disease development
    • C—CHEMISTRY; METALLURGY
    • C12—BIOCHEMISTRY; BEER; SPIRITS; WINE; VINEGAR; MICROBIOLOGY; ENZYMOLOGY; MUTATION OR GENETIC ENGINEERING
    • C12Q—MEASURING OR TESTING PROCESSES INVOLVING ENZYMES, NUCLEIC ACIDS OR MICROORGANISMS; COMPOSITIONS OR TEST PAPERS THEREFOR; PROCESSES OF PREPARING SUCH COMPOSITIONS; CONDITION-RESPONSIVE CONTROL IN MICROBIOLOGICAL OR ENZYMOLOGICAL PROCESSES
    • C12Q2600/00—Oligonucleotides characterized by their use
    • C12Q2600/156—Polymorphic or mutational markers

Definitions

  • This disclosure generally relates to the detection of alterations in a genome, and more specifically to normalizing calculated base level coverage to more effectively detect CNVs.
  • CNVs Copy number variations
  • NGS next-generation sequencing
  • US 2014/0066317 A1 discloses a method for detecting copy number variations that comprises normalizing sequence reads previously partitioned into predefined regions (called “windows” or “bins”).
  • US 2016/0110497 broadly teaches methods, processes, apparatuses and machines for non-invasive assessment of genetic variations.
  • US 2016/0110497 discloses a method for analyzing nucleic acid from a pregnant female with reduced bias (especially for determining the presence or absence of a chromosome trisomy) comprising (a) obtaining counts of sequence reads mapped to portions of a reference genome, which sequence reads are reads of circulating cell-free nucleic acid from a test sample, (b) generating a regression for (i) the counts, and (ii) guanine and cytosine (GC) content, for each of the portions of the reference genome for the test sample, (c) assessing the goodness of fit of the counts and the GC content to a non-linear regression or a linear regression, thereby generating an assessment, (d) normalizing the counts by a process selected according to the assessment, thereby generating normalized counts with reduced bias and (e) analyzing the nucleic acid from the pregnant female according to the normalized counts.
  • Embodiments described herein relate to methods of analyzing sequencing data to detect CNVs in a nucleic acid sample. Detecting CNVs in a nucleic acid sample obtained from a human subject can be informative for determining a presence of cancer in the subject. In one embodiment, detecting CNVs in a nucleic acid sample obtained from a human subject can be used for early detection of cancer in the subject.
  • the methods described herein determines coverage at individual nucleotide bases determined from targeted sequencing reads. Sources of coverage variation can be corrected at the base level. For each gene of a targeted gene panel, the determined base level coverage across bases of the gene can be considered to more effectively detect CNVs of each gene.
  • baseline coverage biases that exist at each base position can be modeled using training data gathered from healthy individuals. Therefore, when analyzing a test sample obtained from a subject, the base level coverage can be determined for each base position in view of the expected coverage biases obtained through modeling. Specifically, if the coverage bias at a base position for a test sample obtained from the subject differs from the expected coverage bias obtained through modeling, coverage biases can be normalized and removed. For a gene in a targeted gene panel, base level coverages across the base positions of the gene are analyzed to determine whether the coverage for the gene differs from an expected level of coverage for the gene as previously determined using training data gathered from healthy individuals. If so, a CNV can be called. The calling of a CNV can indicate a presence of cancer in the subject or that the subject is susceptible to an increased likelihood of developing cancer.
  • the term "individual” refers to a human individual.
  • the term “healthy individual” refers to an individual presumed to not have a cancer or disease.
  • the term “subject” refers to an individual who is known to have, or potentially has, a cancer or disease.
  • sequence reads refers to nucleotide sequences read from a sample obtained from an individual. Generally, sequence reads are obtained post-amplification (e.g., polymerase chain reaction such as bridge amplification) of a nucleic acid fragment that is obtained, or pulled-down from a test sample, using a hybridization probe.
  • sequence reads are obtained post-amplification (e.g., polymerase chain reaction such as bridge amplification) of a nucleic acid fragment that is obtained, or pulled-down from a test sample, using a hybridization probe.
  • cell free nucleic acid refers to nucleic acid fragments that circulate in an individual's body (e.g., bloodstream) and originate from one or more healthy cells and/or from one or more cancer cells.
  • FIG. 1A is an example depiction of the systems used in detecting CNVs in a test sample, in accordance with an embodiment.
  • nucleic acids 140 are isolated from a test sample obtained from a subject (e.g., a patient) and provided to a sequencer 145 for generating sequence reads 150 of the nucleic acids 140.
  • the sequencer 145 is communicatively coupled with one or more computing devices 155. Each computing device 155 can process the sequence reads 150 to determine whether the genome of the subject that the test sample from which the nucleic acids 140 were isolated from includes one or more CNVs.
  • nucleic acids 140 are isolated from a test sample obtained from a subject (e.g., a patient).
  • the test sample may be from a patient known to have or suspected of having cancer.
  • the test sample may be a sample selected from the group consisting of blood, plasma, serum, urine, fecal, and saliva samples.
  • the test sample may comprise a sample selected from the group consisting of whole blood, a blood fraction, a tissue biopsy, pleural fluid, pericardial fluid, cerebral spinal fluid, and peritoneal fluid.
  • the test sample comprises cell-free nucleic acids (e.g., cell-free DNA), wherein the cell-free nucleic acids in the test sample originate from one or more healthy cells and from one or more cancer cells.
  • the nucleic acids 140 undergo enrichment to generate enriched fragments 180.
  • generating enriched fragments includes hybridizing the nucleic acids 140 using hybridization probes and enriching the probe-nucleic acid complex.
  • the enriched fragments 180 are provided to the sequencer 145 for sequencing. Examples of a sequencer 145 for sequencing the enriched fragments 180 from a test sample include the Illumina HiSeq TM or Illumina MiSeq TM Systems. As shown in FIG.
  • the sequencer 145 can include a graphical user interface 148 that enables user interactions with particular tasks (e.g., initiate sequencing, terminate sequencing) as well as one more loading trays 146 for providing the enriched fragment 180 samples and/or necessary buffers for performing the sequencing assays. Therefore, once a user has provided the necessary reagents and enriched fragment 180 samples to the loading trays 146 of the sequencer 145, the user can initiate sequencing by interacting with the graphical user interface 148 of the sequencer 145. The sequencer 145 performs the sequencing and outputs sequence reads 150 of the enriched fragments 180 from the test sample.
  • tasks e.g., initiate sequencing, terminate sequencing
  • the sequencer 145 performs the sequencing and outputs sequence reads 150 of the enriched fragments 180 from the test sample.
  • Each computing device 155 can be one of a personal computer (PC), a desktop computer, a laptop computer, a notebook, a tablet PC, a mobile device.
  • a computing device 155 can be communicatively coupled to the sequencer 145 through a wireless, wired, or a combination of wireless and wired communication technologies.
  • the computing device 155 is configured with a processor and memory storing computer instructions that, when executed by the processor, cause the processor to process the sequence reads 150 and determine the presence of CNVs in the genome of a subject from whom the test sample was obtained.
  • nucleic acids 140 are extracted from a test sample.
  • the test sample is a cfDNA sample obtained from the subject.
  • DNA e.g., cfDNA
  • any known method in the art can be used for purifying DNA.
  • nucleic acids 140 can be isolated by pelleting and/or precipitating the nucleic acids 140 in a tube.
  • a sequencing library is prepared.
  • unique molecular identifiers UMI
  • the UMIs are short nucleic acid sequences (e.g., 4-10 base pairs) that are added to ends of nucleic acids 140 during adapter ligation.
  • UMIs are degenerate base pairs that serve as a unique tag that can be used to identify sequence reads obtained from a nucleic acids. As described later, the UMIs can be further replicated along with the attached nucleic acids during amplification, which provides a way to identify sequence reads that originate from the same original nucleic acid segment in downstream analysis.
  • step 110 in various embodiments, one or more of the steps 115, 120 and 125 are performed by individual hardware devices.
  • a first dedicated device can be used for enrichment 120 of targeted nucleic acids (e.g., informative cell-free DNA fragments).
  • Hybridization probes are hybridized 115 to a set of targeted nucleic acids.
  • the probe-nucleic acid complexes can then be enriched 120 from the sample.
  • a biotin moiety can be added to the 5'-end of the probes (i.e., biotinylated) to facilitate pulling down of target probe-nucleic acids complexes using a streptavidin-coated surface (e.g., streptavidin-coated beads).
  • a second device such as a polymerase chain reaction (PCR) device, can be used for amplification of the targeted nucleic acids.
  • PCR polymerase chain reaction
  • hybridized probe-nucleic acid complexes pulled down from the sample are enriched.
  • FIG. 1C it depicts the processing steps for generating sequence reads of a nucleic acid 140, in accordance with an embodiment.
  • the nucleic acid 140 can have a nucleic acid sequence of "ACTCGACC.”
  • Each nucleotide in the nucleic acid 140 is hereafter referred to as a base position 162.
  • Hybridization probes 165A and 165B can be designed to target and hybridize with the targeted nucleic acid sequence of the nucleic acid 140 to pull down, and enrich, nucleic acid fragments 170 that may be informative for the presence or absence of cancer (or disease), cancer status, or a cancer classification (e.g., cancer type or tissue of origin).
  • a plurality of hybridization pull down probes can be used for a given target sequence or gene.
  • the probes can range in length from about 40 to about 160 bp, from about 60 to about 120 bp, or from about 70 bp to about 100 bp. In one embodiment, the probes cover overlapping portions of the target region or gene.
  • Probe 165A includes a nucleotide sequence (e.g., "TGAGC") that is complementary to a portion of the nucleotide sequence of the targeted nucleic acid 140 (e.g., "ACTCG”). Additionally, probe 165B includes a nucleotide sequence (e.g., "AGCTG”) that is complementary to a portion of the targeted nucleotide sequence of the nucleic acid 140 (e.g., "TCGAC”). Together, probe 165A and probe 165B cover overlapping base positions 162 on the nucleic acid 140 (e.g. "TCG”). Each probe 165A and 165B may include a center position 168A and 168B.
  • the nucleic acid 140 targeted for enrichment with probes 165A and 165B can be used to detect CNVs that may be informative for cancer, which is described in further detail below in relation to FIG. 2 .
  • each of the probes are designed based on a targeted gene panel to analyze particular mutations or target regions of the genome (e.g., of the human or another organism) that are believed to be informative for certain cancers or other types of diseases.
  • a targeted gene panel rather than sequencing all expressed genes of a genome, also known as “whole exome sequencing,” the method can decrease sequencing requirements, or allow for increased sequencing depth, of the target regions.
  • each probe 165A and 165B can be used to pulls down nucleic acid fragments 170 to which they hybridize.
  • each of probe 165A or 165B hybridizes with a portion of the nucleotide sequence of the nucleic acid 140 (e.g., 165A hybridizes with "ACTCG”) and can be enriched.
  • a biotin moiety can be added to the probed to facilitate pulling down of targeted probe-nucleic acids complexes using streptavidin-coated beads.
  • the enriched nucleic acids can be amplified (e.g., using PCR) prior to sequencing.
  • sequence reads 150 are sequenced by the sequencer 145 to generate sequence reads 150.
  • Sequence reads 150 may be acquired by known means in the art.
  • step 125 may employ next generation sequencing (NGS) techniques including synthesis technology, pyrosequencing, ion semiconductor technology, single-molecule real-time sequencing, sequencing by ligation, nanopore sequencing, or paired-end sequencing.
  • NGS next generation sequencing
  • massively parallel sequencing is performed using sequencing-by-synthesis with reversible dye terminators.
  • the sequence reads 150 can be provided by the sequencer 145 to the computing device 155 for further analysis.
  • the sequence reads 150 are provided to the computing device 155 as a BAM file.
  • the computing device 155 may perform steps 130 and 135 to identify CNVs.
  • sequence reads are obtained and aligned to a reference genome.
  • any known method in the art can be used for aligning the sequence reads to a reference genome.
  • the nucleotide bases of a sequence read are aligned with nucleotide bases in the reference genome to determine alignment position information for the sequence read.
  • Alignment position information can include a beginning position and an end position of a region in the reference genome that corresponds to the beginning nucleotide base and end nucleotide base of the sequence read.
  • Alignment position information may also include sequence read length, which can be determined from the beginning position and end position.
  • a BAM file of aligned sequencing reads for regions of the genome is obtained and utilized for analysis in step 135.
  • CNVs are identified using the aligned sequence reads.
  • CNVs are determined for a targeted gene panel to analyze particular mutations or target regions of the genome.
  • coverage at each base position is determined.
  • the coverage at each base position can be susceptible to various biases of that base position such as characteristics of the one or more hybridization probes that were used to determine the base level coverage of the base position, GC content around the base position, and mappability of the base position.
  • the base level coverage for each base position is normalized to account for the various biases.
  • the base level coverage of base positions across the gene for a subject is analyzed to determine whether it deviates from the base level coverage of base positions across the gene for previously analyzed, healthy individuals. If a significant deviation exists, then the gene of the subject likely includes one or more CNVs.
  • the process for identifying CNVs is described in further detail below in reference to FIGs. 2 , 3A , and 3B .
  • FIG. 2 is a flow process for generating models that are applied during deployment for detecting a CNV, in accordance with an embodiment.
  • a first model hereafter referred to as a base level coverage model predicts a base level coverage for each base position of a gene.
  • the base level coverage model takes into consideration the characteristics of one or more probes that are used to pull down a targeted nucleic acid and generate a sequence read that covers the base position.
  • a second model hereafter referred to as a gene level model, determines a likelihood of a CNV of a gene.
  • the gene level model considers the base level coverage across base positions of the gene to determine whether the coverage level across the gene differs from an expected level of coverage of the gene.
  • one or more additional models determine whether an unexpected bias exists for a base position in the genome. Examples of bias can arise from guanine-cytosine content, mappability, general variability, or other unexplained sources.
  • the base coverage correction models may include, but are not limited to, a GC content model, a mappability model, a principal component analysis model, and a base filtering model.
  • observed base level coverage for a base position is denoted as B o ( i ) and theoretical base level coverage for a base position is denoted as B t ( i ).
  • observed scaled base level coverage is denoted as s o ( i ) and theoretical scaled base level coverage is denoted as s t ( i ).
  • an observed base level coverage B o ( i ) for each base position i of the gene is determined using the training sequence reads.
  • an observed base level coverage B o ( i ) for each of a subset of base positions of the gene is determined.
  • the observed base level coverage B o ( i ) for a base position of the gene is the total number of training sequence reads derived from one or more healthy individuals that cover the base position of the gene. Given that healthy individuals are unlikely to have CNVs of genes, the observed base level coverages B o ( i ) of the training sequence reads at each base position can serve as a baseline for determining whether CNVs exist in possible subjects that may be at risk for having or developing cancer.
  • a fit between the observed base level coverage B o ( i ) at the position and locations of hybridization probes used to generate the training sequence reads is performed.
  • FIG. 1C To provide an example of the location of hybridization probes, reference is made to FIG. 1C .
  • the guanine nucleotide base 162 on the nucleic acid 140 is the target base position i for which the fit is being performed.
  • Each probe 165 that hybridizes with bases of the nucleic acid 140 can have a genomic position p c .
  • the genomic position p c of a probe 165 is defined as the position in the genome that aligns with the center position 168 of the probe 165.
  • the genomic position p c of the probe 165A refers to the adenine nucleotide base 168A that is the center nucleotide base of the probe 165A.
  • the genomic position p c of the probe 165A is two nucleotide bases away from the target base position i of the guanine nucleotide base 162.
  • the genomic position p c of the probe 165B refers to the cytosine nucleotide base 168B that is the center nucleotide base of the probe 165B.
  • the genomic position p c of the probe 165B matches the target base position i of the guanine nucleotide base 162.
  • This second degree equation reflects the fact that a probe with a genomic position p c that is near the target base position i heavily contributes to the base level coverage B ( i ) in comparison to a different probe with a genomic position p c that is further from the target base position i .
  • B ( i ) in equation (1) can be replaced with the observed base level coverage B o ( i ) for the base position derived from training sequence reads derived from different healthy individuals.
  • coefficient ⁇ is based, for example, on the overall coverage level of sample coverage and can be considered a constant across all probes.
  • a linear model can be used across all probes and base positions to estimate coefficient ⁇ .
  • the value of the ⁇ coefficient for a probe is based, for example, on the pull-down efficiency of the probe that hybridized with bases within the region of the gene.
  • Probe pull-down efficiency can be determined, for example, based on the guanine-cytosine (GC) content of the probe. Because of their GC content, different probes may have different efficiencies in pulling down a DNA fragment. For example, if a probe has a very high GC content, or a very low GC content, the efficiency in pulling down a DNA fragment is generally low. If a probe has roughly a 50% GC content and 50% AT content then the efficiency is typically higher than probes having higher or lower levels of GC content. In other words, a probe that is more balanced in GC content is more efficient at pulling down DNA fragments than a probe that has extreme values of GC content.
  • GC guanine-cytosine
  • Equation 1 depicts a fit between base coverage B ( i ) of a base position and the genomic coordinate of a single probe.
  • the base level coverage B ( i ) for base position i can be a summation of the individual base level coverages contributed by each probe, where the fit produces parameters ( ⁇ n and ⁇ n ) for each probe.
  • the observed base level coverage B o ( i ) at each base position determined using training sequence reads is used in Equation (2) in place of the general base level coverage B ( i ).
  • theoretical base level coverage B t ( i ) is determined for each base position using the ⁇ n and ⁇ n parameters.
  • the theoretical base level coverage B t ( i ) represents the modeled base level coverage for each base position.
  • the parameters ⁇ n and ⁇ n are now known. Given each probe genomic position p c yields a theoretical base level coverage B t ( i ) at each base position i.
  • the base level coverage model is generated by performing a fit of theoretical base level coverage B t ( i ) across base positions of the gene to obtain parameters of the base level model.
  • the base level coverage model performs a fit of theoretical base level coverage B t ( i ) across a region of the gene, the region of the gene including one or more base positions. Therefore, for a single gene, different parameters of the base level coverage model are determined for each of multiple regions of the gene.
  • the base level coverage model describes a relationship between scaled base level coverage s ( i ) across the base positions of the region of the gene and base level coverage B ( i ).
  • the base level coverage model is a robust linear fit.
  • an observed scaled base level coverage s o ( i ) for each base position is determined using the observed base level coverage B o ( i ) obtained using the training sequence reads.
  • the observed base level coverage for each base position is scaled to the median base level coverage that was observed across the entire sample obtained from a healthy individual. Scaling allows coverages of training sequences derived from different samples to be comparable.
  • the scaled base level coverage s o ( i ) for each base position is determined by dividing the observed base level coverage B o ( i ) by the median base coverage determined for the test sample.
  • the parameters x and y can be determined using Equation (4).
  • the parameters of the base level coverage model e.g., x and y
  • the parameters of the base level coverage model can be stored and subsequently retrieved during deployment for use in determining the presence of a CNV in a sample.
  • base coverage correction models are trained using the training sequence reads of the training examples.
  • Each base coverage correction model considers a level of bias present in the base level coverage at each base position.
  • base coverage correction models include one or more of a GC content model, a mappability model, a principal component analysis model, and a base filtering model.
  • the GC content model predicts a GC bias for a base position based on the level of GC content within a window of the region of the gene.
  • the base position is centered within the window.
  • the size of the window is dependent on the size of the probe or the size of a nucleic acid fragment 170.
  • the size of the window may be larger than the size of the probe or may be larger than the size of the nucleic acid fragment 170.
  • the window of the region of the gene is between 4 and 200 base pairs.
  • the window of the region of the gene is between 5 and 100 base pairs in size.
  • the window of the region of the gene is between 6 and 80 base pairs in size.
  • the window of the region of the gene is between 6 and 80 base pairs in size. In one embodiment, the window of the region of the gene is between 8 and 60 base pairs in size, or any sub-range therein. In another embodiment, the window of the region of the gene is between 10 and 40 base pairs in size. In another embodiment, the window of the region of the gene is between 12 and 30 base pairs in size. In another embodiment, the window of the region of the gene is at least 5 base pairs in size. In another embodiment, the window of the region of the gene is at least 10 base pairs in size. In another embodiment, the window of the region of the gene is at least 15 base pairs in size. In another embodiment, the window of the region of the gene is at least 20 base pairs in size.
  • the window of the region of the gene is at least 25 base pairs in size. In another embodiment, the window of the region of the gene is 21 base pairs in size. In another embodiment, for a base position, the window includes the ten bases to the left and to the right of the base position.
  • the GC content model predicts a GC content for a window around the base position by approximating a smoothing function across the GC content of base positions in the window.
  • the smoothing function is a local polynomial regression (LOESS) curve fit that smooths the GC content across the window. Therefore, the GC content model can output a GC content value by using the approximated smoothing function.
  • LOESS local polynomial regression
  • the GC content model approximates the smoothed function and predicts a baseline GC content for each base position using the smoothed function.
  • the baseline GC content for a base position is the GC content in a window around the base position.
  • the predicted baseline GC content for each base position is stored such that during deployment, the baseline GC content for each base position can be used to determine whether an additional GC correction needs to be performed.
  • the mappability model predicts a mappability bias for a base position based on the mappability of each base position within a window of the region of the gene. Similar to the window described above in reference to the GC content model, the window of the region of the gene used to predict a mappability bias for a base position can be any of the aforementioned sizes. In particular embodiments, the window of the region of the gene used to predict a mappability bias for a base position is 21 base pairs in size
  • the mappability for each base position can be accessed from publicly available databases such as the UC Santa Cruz Genome Browser.
  • the mappability model predicts a mappability bias for a base position by approximating a smoothed function across the mappability of base positions in the window.
  • the smoothing function is a local polynomial regression (LOESS) curve fit that smooths the mappability levels of base positions within the window.
  • the mappability model outputs a mappability for the base position by using the approximated smoothed function.
  • the mappability model predicts a baseline mappability for each base position. The baseline mappability for each base position can be used during deployment to determine whether an additional mappability correction needs to be performed.
  • the principal component analysis model predicts biases for each base position that can arise from unknown sources.
  • the principal component analysis model performs a principal component analysis (PCA) to identify principal components PC n for each observed scaled base level coverage s o ( i ) for each base position i.
  • PCA principal component analysis
  • each of the parameters ( a , b 1 ... b n ) and the principal components PC n are determined using the observed scaled base level coverage s o ( i ) derived from the training examples.
  • These principal components represent the baseline such that during deployment, the principal component analysis model can be applied to determine whether the PCA analysis yields additional unknown biases that differ from the baseline.
  • the base filtering model determines statistical measures of the observed scaled base level coverage s o ( i ) for the base position i. For example, the base filtering model receives, as input, the observed scaled base level coverage s o (i) for each base position and calculates a median base level coverage for each base position and a standard deviation of the base level coverage for each base position. The median and standard deviation of the base level coverage for each base position can serve as a baseline for determining whether a scaled base level coverage determined during deployment is a significant outlier. If so, the scaled base level coverage for the base position can be filtered and removed.
  • a gene level model is generated by performing a fit of base level coverage of positions across multiple regions of the gene to obtain parameters of the gene level model.
  • a is the parameter learned across the training examples and m is the slope of the linear fit, otherwise referred to as the gene coefficient for the gene.
  • the gene level model outputs the gene coefficient for the gene, which can be used to make inferences about the amplification or deletion of regions of the gene.
  • the gene level model can output each gene coefficient m n for each gene in the targeted gene panel to determine the likelihood of a CNV (e.g., duplication or deletion) of a region of the gene.
  • FIG. 3A is a flow diagram of a method for identifying CNVs in a test sample, in accordance with an embodiment.
  • the test sample is processed using the process workflow described in FIG. 1A .
  • the aligned sequence reads of the test sample are generated using processing steps 105-130.
  • FIG. 3A describes step 135 in further detail.
  • the base level coverage for each position in a region of a gene is determined from the aligned sequencing reads that each cover the position.
  • any known method in the art can be used to determine base level coverage.
  • the number of sequence reads overlapping a given base position represents the base level coverage for that base position.
  • the notation B d ( i ) refers to the base level coverage determined, during deployment, for each base position for a target sample obtained from a subject.
  • the notation s d ( i ) refers to the scaled base level coverage determined, during deployment, for each base for the target sample obtained from a subject.
  • the base level coverage B d ( i ) is scaled to the median coverage that was observed within the entire sample.
  • the scaled base coverage is determined by dividing the base level coverage by the median base coverage determined for the test sample. Scaling allows coverages from different samples to be compared.
  • the base level coverage B d ( i ) for each base position is corrected to account for characteristics of the hybridization probes that were used to obtain the nucleic acid fragments (e.g., nucleic acid fragments 170) that were subsequently used to generate the sequence reads.
  • the base level coverage model is retrieved and the base level coverage for each base position within a region of the gene is applied as input to the base level coverage model to determine the scaled base level coverage of each base position.
  • an optional step of correcting for biases at each base position can be performed.
  • FIG. 3B depicts additional steps for correcting biases that arise at each base position, in accordance with an embodiment.
  • Each of the additional steps of performing GC content correction 355, performing mappability correction 360, performing principal component analysis (PCA) 370, and performing base level filtering 375 can be performed independent of another additional correction step.
  • each of steps 355-375 can be performed in different orders.
  • all of steps 355-375 are performed for a sample to correct for biases that may exist at each base position.
  • a subset of steps 355-375 can be performed for a sample to correct for a subset of biases that may exist at each base position.
  • the step of performing a GC content correction includes the application of the GC content model.
  • a window of the region of the gene is identified.
  • the base position is centered within the window.
  • the window of the region of the gene can range from 4 to 200 base pairs, 5 to 100 base pairs, 6 to 80 base pairs, 8 to 60 base pairs, 10 to 40 base pairs, or 12 to 30 base pairs in size.
  • the window of the region of the gene is at least 5 base pairs, 10 base pairs, 15 base pairs, 20 base pairs, 25 base pairs, 30 base pairs, 35 base pairs, or 40 base pairs in size.
  • the window of the region of the gene is 21 base pairs.
  • the GC content of the base positions within the window of the region of the gene is provided as input to the GC content model.
  • the GC content model smooths the GC content across the window of the region of the gene by creating an approximating function.
  • the GC content model performs a local polynomial regression (LOESS) curve fitting to smooth the GC content across the window.
  • LOESS local polynomial regression
  • the GC content model uses the approximated smoothed function to outputs a GC content value for the window.
  • the GC content value for the window is compared to a GC content value for the window that was calculated using the prior training examples.
  • the GC content value for the window of the test sample is compared to a baseline value to determine whether the GC content from the test sample significantly differs from prior healthy samples. If the GC content value differs significantly in comparison to the baseline, an additional GC correction is performed on the scaled base level coverage s d ( i ) determined for the base position.
  • the scaled base level coverage s d ( i ) for the base position can be normalized to the GC content of the window in the target sample.
  • the scaled base level coverage s d (i ) can be modified by a factor representing the increase in GC content in the test sample. This accounts for the less efficient pull down of probes due to the higher, or lower, GC content in the targeted nucleic acid region.
  • performing a mappability correction for a base position includes the application of the mappability model.
  • a window of the region of the gene is identified.
  • the base position is centered within the window.
  • the window of the region of the gene can range from 4 to 200 base pairs, 5 to 100 base pairs, 6 to 80 base pairs, 8 to 60 base pairs, 10 to 40 base pairs, or 12 to 30 base pairs in size.
  • the window of the region of the gene is at least 5 base pairs, 10 base pairs, 15 base pairs, 20 base pairs, 25 base pairs, 30 base pairs, 35 base pairs, or 40 base pairs in size.
  • the window of the region of the gene is 21 base pairs.
  • the mappability at each base position of the window of the region of the gene is identified and provided as input to the mappability model.
  • the mappability model smooths the mappability values of base positions across the window of the region of the gene by creating an approximated smoothed function.
  • the mappability model performs a local polynomial regression (LOESS) curve fitting to smooth the mappability of base positions across the window.
  • the mappability model outputs a mappability value for the base position given the approximated smoothed function. To determine whether an additional mappability correction is needed, the mappability value for the base position is compared to a mappability value for the base position that was calculated using the prior training examples.
  • LOESS local polynomial regression
  • the mappability value for the base position of the test sample is compared to a baseline value to determine whether the mappability value of the base position from the test sample significantly differs from the mappability of the base position from prior healthy samples. If the mappability value differs significantly to the baseline, an additional mappability correction is performed.
  • the scaled base level coverage s d ( i ) for the base position can be normalized relative to the mappability of the base position in the target sample.
  • the principal component analysis model is applied to determine whether an additional PCA correction is needed to account for unknown biases.
  • the base level coverage is provided as input to the PCA model, which determines principal components of the base level coverage.
  • the principal components of the base level coverage are compared to baseline principal components determined using training examples. If the differences between the principal components of the base level coverage differ significantly from the baseline principal components determined using training examples, an additional PCA correction can be performed on the base level coverage s d ( i ) to eliminate additional unknown biases that may be present.
  • base level filtering for each base position can be performed.
  • Base level filtering can be used to remove base positions that are significant outliers prior to calculating gene level coverage summary (e.g., calculating gene coefficients).
  • the standard deviation of base level coverages derived from prior training examples can be used to filter out base positions that significantly deviate from the prior training examples.
  • base positions that possess a base level coverage that differs by at least a threshold amount from the median base coverage m(i) for the base position can be filtered out.
  • the threshold amount is a factor multiplied by the standard deviation var ( i ) of base level coverages derived from prior training examples.
  • the threshold amount is three times the standard deviation var ( i ) of base level coverages derived from prior training examples.
  • the coverage level for a gene is estimated based on the base level coverage across the base positions of the gene.
  • the base level coverage across various base positions can be used to determine a coverage level for a region of the gene.
  • the region of the gene can be evaluated to determine whether a CNV for the region exists.
  • the base level coverage can be the scaled base level coverage s d ( i) determined after step 315 without further corrections (e.g., Step 320). In other embodiments, the base level coverage can be the corrected base level coverage determined after performing additional corrections (e.g., Step 320).
  • the base level coverage of each base position used to calculate gene level coverage at step 325 is hereafter denoted as f d ( i ).
  • the base level coverage of base positions f d ( i ) of the gene are applied as input to the gene level model.
  • the gene level model outputs a gene coefficient, which represents a gene level fold change that can likely be attributed to the presence of a CNV.
  • the gene level model can be a robust linear model that describes a relationship between the base level coverage and base positions of the gene.
  • the linear fit would theoretically yield a gene coefficient of approximately 1.0 if no CNVs of the gene are present.
  • the gene coefficient represents a gene fold change and can be compared to a pre-determined threshold value.
  • the pre-determined threshold value is dependent on instrumentation sensitivity to avoid calling a CNV that may arise due to noise.
  • the pre-determined threshold value is different for each gene of a targeted panel.
  • a pre-determined threshold can be a percentage deviancy from the theoretical gene coefficient of 1.0. As an example, a pre-determined threshold can be an 8% deviancy, which leads to an absolute threshold value of 0.92 or 1.08. If the gene coefficient is less than 0.92, a CNV in the form of a deletion of a region of the gene may be present. Alternatively, if the gene coefficient greater than 1.08, a CNV in the form of a duplication of a region of the gene may be present. Based on the comparison between the gene coefficient and the pre-determined threshold value, a CNV for the gene can be identified.
  • CNVs are detected for a targeted panel of genes.
  • the targeted panel may include 100, 200, 500, or more genes such that a gene level model performs the analysis and determines a gene coefficient representing a gene fold change for each gene in the targeted gene panel. Examples of estimated coverage level for a gene (determined as gene level fold change) are described in further detail below with reference to FIGs. 9A, 9B, 9C , 10A, and 10B .
  • the gene coefficient can be standardized.
  • the gene coefficient is standardized by calculating a standard score.
  • An example of a standard score can be any one of a z-score, normal score, t-score, or standardized variable.
  • the gene coefficient is standardized using gene coefficients g t and a variance var t of the gene coefficients previously determined using prior training examples.
  • the standardized score can be compared to a threshold value to determine whether a CNV is called.
  • the threshold value can be a score dependent on a confidence interval, such as a 90%, 95%, or 99% confidence interval. Therefore, if the standardized score derived from the gene coefficient is larger than the threshold score, a CNV can be called. Conversely, if the standardized score derived from the gene coefficient is less than the threshold score, a CNV is not called.
  • the methods described above, including the method of calling a CNV or detecting presence of cancer in a subject are, in various embodiments, performed on a computer, such as computing device 155 shown in FIG. 1A .
  • the methods described above can be implemented as computer programs executing on programmable computers that include a processor, a data storage system (including volatile and non-volatile memory and/or storage elements), a graphics adapter, a pointing device, a network adapter, at least one input device, and at least one output device.
  • An output device such as a display, can be coupled to the graphics adapter.
  • Program code is applied to input data to perform the functions described above and generate output information.
  • the computer can be, for example, a personal computer, microcomputer, or workstation of conventional design.
  • the methods depicted in FIGs. 2 , 3A , and 3B can be embodied as computer instructions stored on a computer-readable storage medium.
  • the computer instructions when executed by a processor of a computer, such as computing device 155 shown in FIG. 1A , cause the processor to perform the steps of the methods shown in FIGs. 2 , 3A and 3B as described above.
  • the methods of generating the models includes the step of storing the parameters of each model.
  • the parameters of each model can be stored on the computer-readable medium for subsequent retrieval during deployment when one or more of the models are applied to call a CNV in a test sample obtained from a subject.
  • a computer-readable medium that includes the parameters of each model can be received from a third party that generates the models.
  • the parameters of the models can be retrieved and subsequently used to call a CNV in a test sample.
  • a machine-readable storage medium includes machine readable data which, when using a computing device programmed with instructions for using the data, causes the computing device to display results.
  • Example results can include an indication of the existence of a CNV in one or more genes. Such results can be used for a variety of purposes, such as patient monitoring, treatment considerations, and the like.
  • kits performing the methods described above, including the method of calling a CNV or detecting a presence of cancer in a subject.
  • kits can include reagents for isolating nucleic acids from a test sample.
  • the reagents can further include reagents for sequencing the nucleic acids including buffers and detection agents.
  • kits can further include instructions for use of the reagents included in the kit.
  • a kit can include instructions for extracting the nucleic acid from the test sample.
  • Example instructions can be the order in which reagents are to be added, centrifugal speeds to be used to isolate nucleic acids from the test sample, and/or how to operate a hardware device, such as sequencer 145 shown in FIG. 1A , for the purposes of sequencing the nucleic acids of the test sample.
  • kits further include instructions of computer software for calling CNVs.
  • One form in which these instructions can be present is as printed information on a suitable medium or substrate, e.g., a piece or pieces of paper on which the information is printed, in the packaging of the kit, in a package insert.
  • a suitable medium or substrate e.g., a piece or pieces of paper on which the information is printed, in the packaging of the kit, in a package insert.
  • a computer readable medium e.g., diskette, CD, hard-drive, network data storage, on which the instructions have been stored in the form of computer code.
  • Yet another means that can be present is a website address which can be used via the internet to access the information at a removed site.
  • Example 1 Theoretical Base Coverage as a Function of Probes
  • FIG. 4 illustrates a panel of plots of the expected base level coverage as a function of the number and location of probes targeting a given region in a cfDNA sample.
  • the base level coverage 420 increases given an increasing number of probes 410.
  • the positions of probes 410 targeting a region are indicated as lines along the genomic position (e.g., x-axis).
  • a plot of the base coverage 420A is a downward parabola that peaks at about the midpoint of the probe 410A and decreases at base positions further away from the center of the probe 410A.
  • a plot of the base coverage 420B shows a mixed effect where each probe 410B generates its own coverage of a given region; some base positions will be observed using a first probe, some base positions will be observed using a second probe, and for some base positions there will be a cumulative effect.
  • the coverage decreases up to a point where the coverage is near zero. This decrease to zero is due to the relatively short range of values for fragments in a cfDNA sample.
  • the average fragment size for a cfDNA sample is about 170 bp with a fragment size range typically from about 50 bp to about 250 bp.
  • a base location that is 300 bases away from a probe will typically not be pulled down with cfDNA fragments pulled down by the probe and the coverage will be zero.
  • a plot of the base coverage 420C is similar to a plot of the base level coverage 420B for 2 probes except there will be additional areas where there is a mixture of observations generated from each of the 3 probes. Further away from the probes, the coverage will decrease up to a point where the coverage will be zero.
  • Regions targeted with a low number of probes have an average base level coverage that increases with the number of targeting probes. As the number of targeting probes is further increased (e.g., 5 or 6 probes), the base level coverage (e.g., 420D, 420E, and 420F) plateaus and is relatively constant.
  • FIG. 5 is a plot showing the observed scaled base coverage for the TNSFRSF14 gene of a cfDNA sample.
  • the scaled base coverage at each base position of the gene which is represented in FIG. 5 as the gene coordinate (e.g., x-axis), is determined by scaling the base level coverage observed at each base position by the median base level coverage observed across the sample.
  • each vertical, dotted line 575A and 575B delineates a region 560 of the TNSFRSF14 gene from a different region.
  • each region 560 of the gene refers to an exon region whereas intron regions of the gene are not shown.
  • the probes 550 for the region 560 are depicted as a horizontal line, which indicates the position of each probe 550 within the region 560 of the gene.
  • the probes 560 selected for this analysis were probes used for detection of pancreatic cancer.
  • the observed scaled base coverage 510 for each region generally exhibits a downward parabolic shape, as was described above with reference to FIG. 4 .
  • the data shows that for regions 560 targeted by 3 probes, the amplitude of the scaled base coverage 510 is higher compared to a lower amplitude of the scaled base coverage for regions targeted by 2 probes.
  • the parabolic shape of the scaled coverage observe for the TNFRSF 14 gene is also observed for other targeted genes (data not shown). Also depicted in FIG. 5 is the GC content 520 within each region of the TNFRSF14 gene.
  • FIG. 6 is a plot showing the observed scaled base coverage 610 versus scaled base coverage predicted by the base level model 630 using theoretically determined parameters for TNFRSF14 gene in the cfDNA sample of Figure 4 .
  • the vertical dotted lines 675A and 675B each delineates a region 660 of the TNSFRSF14 gene from a different region.
  • the probes 650 for the region 660 are depicted as a horizontal line, which indicates the position of each probe 650 within the region 660 of the gene.
  • the theoretical base level coverage generated by the model is based on two sets of information: the values for ⁇ and ⁇ and the locations of each of the probes.
  • the data shows that the theoretical base level coverage 630 for each region 660 of the gene approximates the observed base coverage 610 for each region 660 of the gene.
  • the difference between the observed base level coverage 610 and theoretical base level coverage 630 is likely due to the theoretical choice of the values for the coefficients ⁇ and ⁇ .
  • FIG. 7 is a plot showing the observed scaled base coverage versus scaled base coverage predicted by a model using parameters derived from training data for the TNFRSF14 gene in the cfDNA sample of FIG. 6 .
  • the vertical dotted lines 775A and 775B each delineates a region 760 of the TNSFRSF14 gene from a different region.
  • the probes 750 for the region 760 are depicted as a horizontal line, which indicates the location of each probe 750 within the region 760 of the gene with respect to the gene coordinate (e.g., x-coordinate).
  • training data including training sequence reads sequenced from a test sample obtained from healthy individuals were used to generate parameters for the base level model (e.g., parameters x and y as described above).
  • parameters for the base level model e.g., parameters x and y as described above.
  • a unique set of parameters for the base level model is generated for each region of the TNFRSF14 gene.
  • the parameter x and y for the base level model for each region were used.
  • the predicted base level coverage 730 represents the scaled base level coverage obtained after performing the coverage correction to account for characteristics of hybridization probes (e.g., step 315 in FIG. 3 ).
  • the predicted scaled base level coverage 730 closely tracks the observed scaled base coverage 710 for each region 760 of the gene. This establishes that the base level model, and the parameters x and y of the base level model, are useful for modeling the base level coverage across base positions of regions 760 of the gene.
  • FIG. 8 is a plot showing the observed scaled base coverage 810 versus the predicted base level coverage 830 for the MET gene in a genomic DNA sample.
  • the genomic DNA sample was prepared by titrating (spiking-in) genomic DNA from a cell line with an amplification of the MET gene into a genomic DNA sample from another cell line with 2 copies of the MET gene.
  • the proportion of the spike-in reflects how much amplification is expected for the MET gene.
  • the expected amplification for MET is 1.4x fold change.
  • training data including sequence reads sequenced from a test sample obtained from healthy individuals were used to generate parameters for the base level model (e.g., parameters x and y as described above).
  • the scaled base level coverage 830 predicted by the base level model at each base position across various regions of the MET gene closely tracks the observed base level coverage 810 at each base position. This validates the parameters for the base level model. Additionally, as shown in FIG. 8 , the amplitudes (measured on the y-axis) of both the observed base level coverage 810 and the scaled base level coverage 830 are in concordance with the expected amplification of 1.4x fold change.
  • genomic DNA samples were prepared by titrating (spiking-in) genomic DNA from a cell line with amplification of the MET gene into a genomic DNA sample from another cell line with 2 copies of the MET gene.
  • genomic DNA samples with expected 1.18x and 1.08x amplifications of MET were also prepared.
  • the predicted scaled base level coverage for each of the genomic DNA samples with expected 1.18x and 1.08x amplifications also tracked closely to the observed base level coverage.
  • the scaled base level coverage (e.g., 830 shown in FIG. 8 ) across the base positions of the MET gene were applied as input to the gene level model, which is a robust linear model (e.g., Equation (9)) that outputs a MET gene coefficient.
  • the gene coefficient for each genomic DNA sample represents the estimated gene fold change of the MET gene in the sample.
  • FIG. 9A is a plot showing the estimated gene fold change in a genomic DNA sample with an expected 1.4x amplification of the MET gene. Each dot in the plot represents a gene with a copy number of 2.
  • the parameter a of the gene level model is previously determined based on training data that includes sequence reads sequenced from samples obtained from healthy individuals.
  • the MET gene coefficient 950A predicted by the gene level model exhibits a 1.4x amplification and is in concordance with the expected results.
  • a similar analysis was conducted to determine a gene coefficient for the additional gene.
  • Training data including sequence reads derived from healthy individuals were used to generate the parameter (e.g., parameter a ) for a robust linear model for the additional gene.
  • the scaled base level coverage across regions of each additional gene is applied as input to the gene level model to determine the gene coefficient for the additional gene.
  • a large majority of gene coefficients for the additional genes fall below a threshold value.
  • the threshold value is set to ensure that gene coefficients above the threshold value 930A or below the threshold value 930B are unlikely to have arisen due to random noise.
  • FIG. 9B is a plot showing the estimated gene fold change in a genomic DNA sample with an expected 1.18x amplification of the MET gene.
  • FIG. 9C is a plot showing the estimated gene fold change in a genomic DNA sample with an expected 1.08x amplification of the MET gene.
  • the respective MET gene coefficient 950B and 950C is plotted.
  • the MET gene coefficient 950B is approximately 1.18, which aligns with the expected amplification.
  • the MET gene coefficient 950C is approximately 1.08, which also aligns with the expected amplification.
  • 9C establishes that the estimated fold change for the MET gene is concordant with the expected 1.08x amplification for the MET gene; however the level of fold change is within the noise level in the data from additional genes.
  • the limit of detection is approximately 1.08x amplification.
  • FIG. 10A is a plot showing the relationship between the expected fold change (x-axis) and the estimated fold change (y-axis) for the MET gene in the genomic DNA samples of FIG. 9A-9C .
  • Figure 10A is a plot showing the relationship between the expected fold change (x-axis) and the estimated fold change in the fragmented genomic DNA samples (y-axis) for the MET gene. As shown in Figure 10A , good correlation is observed between the expected fold change and the estimated fold change for MET amplification in the genomic DNA samples.
  • FIG. 10B is a plot showing the relationship between the expected fold change (x-axis) and the estimated fold change (y-axis) for the ERBB2 gene in genomic DNA samples.
  • the genomic DNA samples were prepared by titrating (spiking-in) genomic DNA from a cell line with amplification of the ERBB2 gene into a genomic DNA sample from another cell line with 2 copies of the ERBB2 gene.
  • Genomic DNA samples were prepared with expected 1.4x, 1.20x, 1.10x, and 1.04x amplifications of the ERBB2 gene and validated by digital PCR. As shown in Figure 10B , good correlation is observed between the expected fold change and the estimated fold change for the ERBB2 gene amplification in the genomic DNA samples.
  • FIG. 11 is a plot showing an ROC curve that indicates detecting 1.08x amplifications or higher in cfDNA samples.
  • the ROC curve was generated using a list of 42 genes known to have CNVs in various types of cancer, measured across 28 samples out of which 12 genes have a 1.08x amplification or higher. Z-scores were generated for all genes and different cut-offs were used for generating the ROC curve.

Landscapes

  • Life Sciences & Earth Sciences (AREA)
  • Chemical & Material Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Proteomics, Peptides & Aminoacids (AREA)
  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Organic Chemistry (AREA)
  • Analytical Chemistry (AREA)
  • Biotechnology (AREA)
  • General Health & Medical Sciences (AREA)
  • Biophysics (AREA)
  • Genetics & Genomics (AREA)
  • Wood Science & Technology (AREA)
  • Zoology (AREA)
  • Medical Informatics (AREA)
  • Molecular Biology (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Theoretical Computer Science (AREA)
  • Spectroscopy & Molecular Physics (AREA)
  • Evolutionary Biology (AREA)
  • Immunology (AREA)
  • Microbiology (AREA)
  • General Engineering & Computer Science (AREA)
  • Biochemistry (AREA)
  • Pathology (AREA)
  • Oncology (AREA)
  • Hospice & Palliative Care (AREA)
  • Epidemiology (AREA)
  • Public Health (AREA)
  • Evolutionary Computation (AREA)
  • Databases & Information Systems (AREA)
  • Data Mining & Analysis (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Bioethics (AREA)
  • Artificial Intelligence (AREA)
  • Software Systems (AREA)
  • Measuring Or Testing Involving Enzymes Or Micro-Organisms (AREA)
  • Chemical Kinetics & Catalysis (AREA)
  • Algebra (AREA)

Claims (15)

  1. Procédé mis en œuvre par ordinateur pour détecter des variations du nombre de copies dans un échantillon, le procédé comprenant les étapes suivantes :
    obtenir des lectures de séquences pour une pluralité de fragments d'acide nucléique, la pluralité de fragments d'acide nucléique étant obtenue à partir de l'échantillon à l'aide d'une pluralité de sondes ;
    pour chaque position de base d'un gène :
    déterminer la couverture au niveau des bases de la position de base à partir des lectures de séquences qui couvrent la position de base ;
    modifier la couverture au niveau des bases de la position afin de tenir compte des caractéristiques de chacune d'un sous-ensemble de la pluralité de sondes qui ciblent une partie du gène qui couvre la position de base ; et
    appliquer une ou plusieurs couvertures de niveau de base modifiées pour une pluralité de positions de base du gène en tant qu'entrée à un modèle linéaire afin de déterminer un coefficient de gène indiquant un niveau de couverture à travers le gène par rapport à un niveau de couverture du gène dans des échantillons d'apprentissage ;
    où la modification de la couverture au niveau des bases de la position de base comprend l'application d'un premier modèle, ledit premier modèle étant généré par les étapes suivantes :
    obtenir des exemples d'apprentissage comprenant des lectures de séquences d'apprentissage qui croisent chacune au moins l'une de la pluralité de positions de base d'une région du gène ;
    déterminer la couverture au niveau des bases pour chacune de la pluralité de positions de base du gène sur la base des lectures de séquences d'apprentissage qui croisent la position de base du gène ; et
    réaliser un ou plusieurs ajustements de la couverture au niveau des bases des lectures de séquences d'apprentissage à travers la pluralité de positions de base du gène afin d'obtenir un ou plusieurs paramètres du premier modèle ;
    où la réalisation desdits un ou plusieurs ajustements de la couverture au niveau des bases comprend en outre les étapes suivantes :
    réaliser un ajustement polynomial de la couverture au niveau des bases sur la pluralité de positions de base de la région du gène afin d'obtenir un premier ensemble de paramètres ;
    pour chaque position de base dans la région du gène, déterminer une couverture théorique au niveau des bases à l'aide du premier ensemble de paramètres ; et
    générer des paramètres du premier modèle en réalisant un ajustement linéaire des couvertures théoriques au niveau des bases sur la pluralité de positions de base de la région du gène ;
    où une ou plusieurs des caractéristiques d'une sonde utilisée pour obtenir un fragment d'acide nucléique sont représentées comme un paramètre du premier modèle ;
    et où les caractéristiques d'une sonde utilisée pour obtenir un fragment d'acide nucléique comprennent :
    (i) une efficacité de capture, ou pull-down, de la sonde ; et/ou
    (ii) une distance entre une coordonnée génomique du gène sur lequel la sonde est centrée et la position de base du gène ; et/ou
    (iii) une teneur en GC dans la sonde.
  2. Procédé selon la revendication 1, comprenant en outre l'appel d'une variation du nombre de copies pour le gène en comparant le coefficient déterminé du gène à une valeur seuil.
  3. Procédé selon la revendication 2, comprenant en outre la détermination de la présence d'un cancer dans l'échantillon sur la base de l'appel de la variation du nombre de copies.
  4. Procédé selon l'une quelconque des revendications 1 à 3, dans lequel la modification de la couverture au niveau des bases de la position de base comprend les étapes suivantes :
    (i) redimensionner la couverture au niveau des bases de la position de base à l'aide d'un niveau de couverture au niveau des bases médian de l'échantillon ; et/ou
    (ii) normaliser la couverture au niveau des bases pour la teneur en GC dans une région du gène, où, en option, la modification de la couverture au niveau des bases pour la teneur en GC dans la région du gène comprend la réalisation d'un ajustement sur la teneur en GC dans une fenêtre de la région du gène, et, éventuellement, où la fenêtre de la région du gène a une taille comprise entre 4 et 200 paires de bases, entre 5 et 100 paires de bases, entre 6 et 80 paires de bases, entre 8 et 60 paires de bases, entre 10 et 40 paires de bases, entre 12 et 30 paires de bases, au moins 5 paires de bases, au moins 10 paires de bases, au moins 15 paires de bases, au moins 20 paires de bases, au moins 25 paires de bases, au moins 30 paires de bases, au moins 35 paires de bases, au moins 40 paires de bases ou 21 paires de bases ; et/ou
    (iii) normaliser la couverture au niveau des bases pour un biais de mappabilité résultant d'une mappabilité de la position de base, où, en option, la normalisation de la couverture au niveau des bases pour un biais de mappabilité comprend la réalisation d'un ajustement sur la mappabilité des positions de base dans une fenêtre de la région du gène ; et/ou
    (iv) réaliser une analyse en composantes principales pour identifier les biais provenant de sources inconnues, où, en option, les composantes principales de l'analyse en composantes principales sont déterminées pour la position de base à partir d'échantillons d'apprentissage.
  5. Procédé selon l'une quelconque des revendications 1 à 4, comprenant en outre les étapes suivantes :
    avant de déterminer le niveau de couverture pour le gène, filtrer une ou plusieurs positions du gène, le filtrage comprenant :
    la détermination d'un niveau de couverture médian du gène sur la base des niveaux de couverture déterminés des positions de base du gène,
    l'élimination des positions de base du gène ayant un niveau de couverture qui diffère du niveau de couverture médian au-delà d'une quantité seuil ;
    et, en option, où la quantité seuil dépend d'un écart de niveau de couverture du gène calculé à partir des niveaux de couverture déterminés des positions de base du gène.
  6. Procédé selon l'une quelconque des revendications 1 à 5, dans lequel l'échantillon d'essai est enrichi en fragments d'ADN indicatifs d'un ou plusieurs des éléments suivants : la présence ou l'absence d'un cancer, le statut d'un cancer ou une classification d'un cancer.
  7. Procédé selon l'une quelconque des revendications 1 à 6, dans lequel les lectures de séquences obtenues sont séquencées à partir d'un échantillon d'acide nucléique exempt de cellules obtenu à partir d'un individu, le procédé comprenant en outre, de manière optionnelle, les étapes suivantes :
    collecter ou avoir préalablement collecté l'échantillon d'acide nucléique exempt de cellules à partir d'un échantillon de sang de l'individu ; et
    enrichir la pluralité de fragments d'acide nucléique provenant de l'échantillon d'acide nucléique exempt de cellules afin de générer les lectures de séquences.
  8. Procédé selon l'une quelconque des revendications 1 à 7, dans lequel les lectures de séquence obtenues sont séquencées à partir d'un échantillon de sang, de sang total, de plasma, de sérum, d'urine, de liquide céphalo-rachidien, de matières fécales, de salive, de larmes, d'une biopsie tissulaire, de liquide pleural, de liquide péricardique ou de liquide péritonéal d'un individu.
  9. Procédé selon l'une quelconque des revendications 1 à 8, dans lequel le gène est inclus dans un panel de gènes ciblés.
  10. Système pour détecter des variations du nombre de copies dans un échantillon, le système comprenant :
    une mémoire de stockage pour stocker des lectures de séquences pour une pluralité de fragments d'acide nucléique, la pluralité de fragments d'acide nucléique étant obtenue à partir d'un échantillon à l'aide d'une pluralité de sondes ;
    un processeur couplé de manière communicative à la mémoire de stockage pour appeler une variation du nombre de copies d'un gène, où l'appel de la variation du nombre de copies du gène comprend les étapes suivantes :
    pour chaque position de base du gène :
    déterminer la couverture au niveau des bases de la position de base à partir des lectures de séquences qui couvrent la position de base ;
    modifier la couverture au niveau des bases de la position afin de tenir compte des caractéristiques de chacune d'un sous-ensemble de la pluralité de sondes qui ciblent une partie du gène qui couvre la position de base ; et
    appliquer une ou plusieurs couvertures de niveau de base modifiées pour une pluralité de positions de base du gène en tant qu'entrée à un modèle linéaire afin de déterminer un coefficient de gène indiquant un niveau de couverture à travers le gène par rapport à un niveau de couverture du gène dans des échantillons d'apprentissage ;
    où la modification de la couverture au niveau des bases de la position de base comprend l'application d'un premier modèle, ledit premier modèle étant généré par les étapes suivantes :
    obtenir des exemples d'apprentissage comprenant des lectures de séquences d'apprentissage qui croisent chacune au moins l'une de la pluralité de positions de base d'une région du gène ;
    déterminer la couverture au niveau des bases pour chacune de la pluralité de positions de base du gène sur la base des lectures de séquences d'apprentissage qui croisent la position de base du gène ; et
    réaliser un ou plusieurs ajustements de la couverture au niveau des bases des lectures de séquences d'apprentissage à travers la pluralité de positions de base du gène afin d'obtenir un ou plusieurs paramètres du premier modèle ;
    où la réalisation desdits un ou plusieurs ajustements de la couverture au niveau des bases comprend en outre les étapes suivantes :
    réaliser un ajustement polynomial de la couverture au niveau des bases sur la pluralité de positions de base de la région du gène afin d'obtenir un premier ensemble de paramètres ;
    pour chaque position de base dans la région du gène, déterminer une couverture théorique au niveau des bases à l'aide du premier ensemble de paramètres ; et
    générer des paramètres du premier modèle en réalisant un ajustement linéaire des couvertures théoriques au niveau des bases sur la pluralité de positions de base de la région du gène ;
    où une ou plusieurs des caractéristiques d'une sonde utilisée pour obtenir un fragment d'acide nucléique sont représentées comme un paramètre du premier modèle ;
    et où les caractéristiques d'une sonde utilisée pour obtenir un fragment d'acide nucléique comprennent :
    (i) une efficacité de capture, ou pull-down, de la sonde ; et/ou
    (ii) une distance entre une coordonnée génomique du gène sur lequel la sonde est centrée et la position de base du gène ; et/ou
    (iii) une teneur en GC dans la sonde.
  11. Système selon la revendication 10, dans lequel l'appel de la variation du nombre de copies pour le gène est effectué en comparant le coefficient déterminé du gène à une valeur seuil.
  12. Système selon la revendication 11, comprenant en outre de déterminer la présence d'un cancer dans l'échantillon sur la base de l'appel de la variation du nombre de copies.
  13. Support de stockage non transitoire lisible par ordinateur stockant un code de programme exécutable par ordinateur qui, lorsqu'il est exécuté par un processeur, amène le processeur à traiter des lectures de séquences pour une pluralité de fragments d'acide nucléique, la pluralité de fragments d'acide nucléique étant obtenue à partir d'un échantillon à l'aide d'une pluralité de sondes, afin de :
    pour chaque position de base d'un gène :
    déterminer la couverture au niveau des bases de la position de base à partir des lectures de séquences qui couvrent la position de base du gène ;
    modifier la couverture au niveau des bases de la position de base afin de tenir compte des caractéristiques de chacune d'un sous-ensemble de la pluralité de sondes qui ciblent une région du gène qui couvre la position de base ; et
    appliquer une ou plusieurs couvertures de niveau de base modifiées pour une pluralité de positions de base du gène en tant qu'entrée à un modèle linéaire afin de déterminer un coefficient de gène indiquant un niveau de couverture à travers le gène par rapport à un niveau de couverture du gène dans des échantillons d'apprentissage ;
    où la modification de la couverture au niveau des bases de la position de base comprend l'application d'un premier modèle, ledit premier modèle étant généré par les étapes suivantes :
    obtenir des exemples d'apprentissage comprenant des lectures de séquences d'apprentissage qui croisent chacune au moins l'une de la pluralité de positions de base d'une région du gène ;
    déterminer la couverture au niveau des bases pour chacune de la pluralité de positions de base du gène sur la base des lectures de séquences d'apprentissage qui croisent la position de base du gène ; et
    réaliser un ou plusieurs ajustements de la couverture au niveau des bases des lectures de séquences d'apprentissage à travers la pluralité de positions de base du gène afin d'obtenir un ou plusieurs paramètres du premier modèle ;
    où la réalisation desdits un ou plusieurs ajustements de la couverture au niveau des bases comprend en outre les étapes suivantes :
    réaliser un ajustement polynomial de la couverture au niveau des bases sur la pluralité de positions de base de la région du gène afin d'obtenir un premier ensemble de paramètres ;
    pour chaque position de base dans la région du gène, déterminer une couverture théorique au niveau des bases à l'aide du premier ensemble de paramètres ; et
    générer des paramètres du premier modèle en réalisant un ajustement linéaire des couvertures théoriques au niveau des bases sur la pluralité de positions de base de la région du gène ;
    où une ou plusieurs des caractéristiques d'une sonde utilisée pour obtenir un fragment d'acide nucléique sont représentées comme un paramètre du premier modèle ;
    et où les caractéristiques d'une sonde utilisée pour obtenir un fragment d'acide nucléique comprennent :
    (i) une efficacité de capture, ou pull-down, de la sonde ; et/ou
    (ii) une distance entre une coordonnée génomique du gène sur lequel la sonde est centrée et la position de base du gène ; et/ou
    (iii) une teneur en GC dans la sonde.
  14. Support de stockage non transitoire lisible par ordinateur selon la revendication 13, dans lequel le traitement des lectures de séquences comprend en outre l'appel d'une variation du nombre de copies pour le gène en comparant le coefficient déterminé du gène à une valeur seuil.
  15. Support de stockage non transitoire lisible par ordinateur selon la revendication 14, dans lequel le traitement des lectures de séquences comprend en outre de déterminer la présence d'un cancer dans l'échantillon sur la base de l'appel de la variation du nombre de copies.
EP17838107.5A 2016-12-22 2017-12-22 Normalisation de couverture de base et son utilisation pour détecter une variation du nombre de copies Active EP3559841B1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US201662438348P 2016-12-22 2016-12-22
PCT/US2017/068305 WO2018119438A1 (fr) 2016-12-22 2017-12-22 Normalisation de couverture de base et son utilisation pour détecter une variation du nombre de copies

Publications (2)

Publication Number Publication Date
EP3559841A1 EP3559841A1 (fr) 2019-10-30
EP3559841B1 true EP3559841B1 (fr) 2025-08-06

Family

ID=61157285

Family Applications (1)

Application Number Title Priority Date Filing Date
EP17838107.5A Active EP3559841B1 (fr) 2016-12-22 2017-12-22 Normalisation de couverture de base et son utilisation pour détecter une variation du nombre de copies

Country Status (3)

Country Link
US (2) US12100483B2 (fr)
EP (1) EP3559841B1 (fr)
WO (1) WO2018119438A1 (fr)

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103748236B (zh) 2011-04-15 2018-12-25 约翰·霍普金斯大学 安全测序系统
AU2013338393C1 (en) 2012-10-29 2024-07-25 The Johns Hopkins University Papanicolaou test for ovarian and endometrial cancers
WO2017027653A1 (fr) 2015-08-11 2017-02-16 The Johns Hopkins University Analyse du fluide d'un kyste ovarien
CN111868260B (zh) 2017-08-07 2025-02-21 约翰斯霍普金斯大学 用于评估和治疗癌症的方法和材料
CA3170345A1 (fr) 2020-02-14 2021-08-19 The Johns Hopkins University Methodes et materiels d'evaluation d'acides nucleiques
WO2024192121A1 (fr) * 2023-03-13 2024-09-19 Grail, Llc Détection d'une contamination par des globules blancs

Family Cites Families (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP2245198A1 (fr) 2008-02-04 2010-11-03 Massachusetts Institute of Technology Sélection d'acides nucléiques par hybridation en solution en amorces oligonucléotidiques
EP2406400B1 (fr) 2009-03-09 2016-07-13 Life Technologies Corporation Méthodes de détermination du nombre de copies d'une séquence de génome dans un échantillon biologique
US20140066317A1 (en) 2012-09-04 2014-03-06 Guardant Health, Inc. Systems and methods to detect rare mutations and copy number variation
CN104603284B (zh) 2012-09-12 2016-08-24 深圳华大基因研究院 利用基因组测序片段检测拷贝数变异的方法
EP4495936A3 (fr) 2013-03-06 2025-08-06 Life Technologies Corporation Systèmes et procédés de détermination de variation de nombre de copies
US10930368B2 (en) 2013-04-03 2021-02-23 Sequenom, Inc. Methods and processes for non-invasive assessment of genetic variations
EP3965111B8 (fr) 2013-08-30 2025-06-11 Personalis, Inc. Méthodes d'analyse génomique
KR102373647B1 (ko) 2013-10-21 2022-03-11 베리나타 헬스, 인코포레이티드 사본수 변동을 결정함에 있어서 검출의 감수성을 향상시키기 위한 방법
GB201319779D0 (en) 2013-11-08 2013-12-25 Cartagenia N V Genetic analysis method
EP3502273B1 (fr) 2014-12-12 2020-07-08 Verinata Health, Inc. Fragment d'adn sans cellules
CN108603228B (zh) 2015-12-17 2023-09-01 夸登特健康公司 通过分析无细胞dna确定肿瘤基因拷贝数的方法
US10095831B2 (en) 2016-02-03 2018-10-09 Verinata Health, Inc. Using cell-free DNA fragment size to determine copy number variations

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
ANDY S. ALIC ET AL: "Objective review of de novo stand-alone error correction methods for NGS data : Objective review of de novo stand-alone NGS correctors", WILEY INTERDISCIPLINARY REVIEWS: COMPUTATIONAL MOLECULAR SCIENCE, vol. 6, no. 2, 11 January 2016 (2016-01-11), pages 111 - 146, XP055449573, ISSN: 1759-0876, DOI: 10.1002/wcms.1239 *
XIE CHAO ET AL: "CNV-seq, a new method to detect copy number variation using high-throughput sequencing", BMC BIOINFORMATICS, BIOMED CENTRAL , LONDON, GB, vol. 10, no. 1, 6 March 2009 (2009-03-06), pages 80, XP021047346, ISSN: 1471-2105, DOI: 10.1186/1471-2105-10-80 *

Also Published As

Publication number Publication date
US20240387000A1 (en) 2024-11-21
US20180225413A1 (en) 2018-08-09
EP3559841A1 (fr) 2019-10-30
WO2018119438A1 (fr) 2018-06-28
US12100483B2 (en) 2024-09-24

Similar Documents

Publication Publication Date Title
JP7667825B2 (ja) 染色体提示の決定
EP3559841B1 (fr) Normalisation de couverture de base et son utilisation pour détecter une variation du nombre de copies
US20240321389A1 (en) Models for Targeted Sequencing
EP3464626B1 (fr) Méthodes de détection de variations génétiques
JP6971845B2 (ja) 遺伝子の変動の非侵襲的評価のための方法および処理
TWI636255B (zh) 癌症檢測之血漿ｄｎａ突變分析
CN112575075B (zh) 遗传变异的非侵入性评估方法和过程
CN112020565A (zh) 用于确保基于测序的测定的有效性的质量控制模板
EP3378001B1 (fr) Procédés pour détecter des variations du nombre de copies dans un séquençage de nouvelle génération
US20210065842A1 (en) Systems and methods for determining tumor fraction
US20200340064A1 (en) Systems and methods for tumor fraction estimation from small variants
CA3121926A1 (fr) Systemes et procedes d'estimation de fractions de source cellulaire a l'aide d'informations de methylation
US20210238668A1 (en) Biterminal dna fragment types in cell-free samples and uses thereof
US20190287646A1 (en) Identifying copy number aberrations
HK40012222B (en) Chromosome representation determinations
BR112020013636B1 (pt) Método para analisar uma amostra materna obtida de uma gestante para facilitar determinação pré-natal da presença de um distúrbio genético, método para identificação de contaminação associada a pelo menos um entre preparação de biblioteca de sequenciamento e sequenciamento de alto rendimento, e método para caracterização associada a pelo menos um entre preparação de biblioteca de sequenciamento e sequenciamento
HK1234182A1 (en) Chromosome representation determinations

Legal Events

Date Code Title Description
STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: UNKNOWN

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: THE INTERNATIONAL PUBLICATION HAS BEEN MADE

PUAI Public reference made under article 153(3) epc to a published international application that has entered the european phase

Free format text: ORIGINAL CODE: 0009012

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: REQUEST FOR EXAMINATION WAS MADE

17P Request for examination filed

Effective date: 20190718

AK Designated contracting states

Kind code of ref document: A1

Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR

AX Request for extension of the european patent

Extension state: BA ME

RIN1 Information on inventor provided before grant (corrected)

Inventor name: BARBACIORU, CATALIN

DAV Request for validation of the european patent (deleted)
DAX Request for extension of the european patent (deleted)
RAP1 Party data changed (applicant data changed or rights of an application transferred)

Owner name: GRAIL, LLC

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: EXAMINATION IS IN PROGRESS

17Q First examination report despatched

Effective date: 20220520

P01 Opt-out of the competence of the unified patent court (upc) registered

Effective date: 20230506

RAP3 Party data changed (applicant data changed or rights of an application transferred)

Owner name: GRAIL, INC.

REG Reference to a national code

Ref country code: DE

Ref legal event code: R079

Free format text: PREVIOUS MAIN CLASS: G06F0019220000

Ipc: C12Q0001686900

Ref document number: 602017091067

Country of ref document: DE

GRAP Despatch of communication of intention to grant a patent

Free format text: ORIGINAL CODE: EPIDOSNIGR1

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: GRANT OF PATENT IS INTENDED

RIC1 Information provided on ipc code assigned before grant

Ipc: G16B 30/10 20190101ALI20250210BHEP

Ipc: G16B 30/00 20190101ALI20250210BHEP

Ipc: G16B 20/20 20190101ALI20250210BHEP

Ipc: G16B 20/00 20190101ALI20250210BHEP

Ipc: C12Q 1/6886 20180101ALI20250210BHEP

Ipc: C12Q 1/6869 20180101AFI20250210BHEP

INTG Intention to grant announced

Effective date: 20250303

GRAS Grant fee paid

Free format text: ORIGINAL CODE: EPIDOSNIGR3

GRAA (expected) grant

Free format text: ORIGINAL CODE: 0009210

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: THE PATENT HAS BEEN GRANTED

AK Designated contracting states

Kind code of ref document: B1

Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR

REG Reference to a national code

Ref country code: GB

Ref legal event code: FG4D

REG Reference to a national code

Ref country code: CH

Ref legal event code: EP

REG Reference to a national code

Ref country code: DE

Ref legal event code: R096

Ref document number: 602017091067

Country of ref document: DE

REG Reference to a national code

Ref country code: IE

Ref legal event code: FG4D